hey squad
shall we settle in?

01 · STR · 03/04

E-commerce strategy.

Not a Shopify checklist. Not a promo roadmap. A loop driven first by margin.

You set your gross margin and target lifetime value, we deliver an acquisition, retention and basket mechanic calibrated on your acceptable CAC. LTV conditions CAC, never the other way around.

By Jérôme · Strategy & data

Get my strategy quotedSee the Ecostal mission

02 · OPERATION NOTE · OPS E-COMMERCE STRATEGY

E-commerce strategy, decoded

The field report

Your CACthe cost to acquire a new customer climbs, your average basket stalls, and retention no longer compensates. Or you are launching an eshop on a market you know offline but not online, and you want to avoid funding 12 months of Meta + Google learning.

The cost of waiting

Without a written loop, you steer acquisition, retention and basket as three separate projects. Arbitrages happen in silos, and net margin dilutes. What it really costs you is the customer value over 12-24 months (LTVthe customer value over 12-24 months) no ad campaign recovers later.

In black and white

A written LTVthe customer value over 12-24 months → CACthe cost to acquire a new customer → basket mechanic, indexed on your gross margin. A per-phase lever grid (acquisition, first purchase, retention, reactivation). A frame to say no to promos that destroy margin without serving loyalty.

The manoeuvre

We look at 18 months of customer-cohortclient groups by acquisition month data, before laying down a single lever. We extract that data into BigQuerythe Google dedicated data warehouse and turn it into Looker StudioGoogle reporting dashboards dashboards, cross-checked with your GA4 Enhanced E-commerce. Hypotheses validated against your gross margin, scenarios delivered as a quarterly Notion grid.

The lock that breaks

E-commerce arbitrages defendable at executive committee. A ground ready to plug in marketing automation on clean LTVthe customer value over 12-24 months cohorts. A clean base for AI workflows downstream, free of last-clickwhere the last click takes all the credit bias.

03 · WHAT WE WON'T WRITE IN AN RFP

An e-commerce strategy is only useful if lifetime value conditions acquisition cost.

Most strategies we inherit reason backwards: how many leads can I buy, and we'll see after. The result: a climbing CAC, a retention no campaign feeds, and a net margin draining. The right sequence: LTV (customer value over 12-24 months) first, acceptable CAC second, acquisition levers third. The field confirms it: across the 579 Liège e-shops we x-rayed, the median commercial maturity index sits between 0.93 and 1.08 across segments, an entire fabric starting from roughly the same place. If the measurement mechanic does not hold, the agency vs freelance vs in-house debate stays theoretical.

  • 01

    LTV computed per cohort

    You know the real customer value at 12, 24, 36 months by entry channel. No more global average hiding unprofitable ad cohorts (clients won via paid ads) beneath very profitable organic ones.

  • 02

    Acceptable CAC per segment

    Each catalog segment has its CAC ceiling, indexed on gross margin and LTV. You no longer over-invest in promos that bring volume without margin, or low-value-over-time customers.

  • 03

    Quantified retention loop

    Your repurchase and reactivation plan is tied to measured LTV cohorts. Each loyalty scenario has an estimated cost and return up front, not validated after the fact.

  • 04

    Filter against margin-killing promos

    You receive six promo proposals a month. The written frame gives you the criterion to say yes or no in 5 minutes. Promos that destroy margin without feeding retention drop off the calendar.

04 · THE MISSION THAT PROVED IT

Three missions where we held it.

Ecostal Group team handling a solar panel in their warehouse

Ecostal Group

B2B photovoltaic e-commerce

18-month e-commerce strategy across three eshops.

Read the Ecostal mission
Brasero estate, corporate retreats and weddings in Liège province

Brasero Group

Hospitality & events · online bookings

24-month LTV/CAC strategy.

Read the Brasero mission
Battlekart, immersive karting in Verlaine and Malmedy

Battlekart

Indoor leisure · online ticketing

18-month purchase journey architecture.

Read the Battlekart mission

05 · THE PLAY-BY-PLAY

Four steps. Six weeks on average. No promo launched without margin calculation.

  1. STEP 01

    We audit 18 months of LTV cohorts.

    Reading by customer cohorts (by entry channel, by catalog segment, by acquisition month). Data extracted into BigQuery (Google dedicated data warehouse) wired to your GA4 Enhanced E-commerce + your ERP or Shopify. We measure real LTV at 12 and 24 months, not the theoretical LTV from Meta / Google dashboards.

  2. STEP 02

    We set the CAC ceiling per segment.

    Workshop with your CFO and e-commerce lead. Per-segment gross margin against measured LTV. CAC ceiling signed in writing. Segments that cannot sustain a viable CAC drop out of the ad scope, the arbitrage is documented.

  3. STEP 03

    We model the retention loop.

    Per-cohort repurchase plan, reactivation window calibrated on natural purchase frequency, basket scenarios via cross-sell and up-sell (related products and upsell). Looker Studio wires the threshold indicators. The plan doesn't ship until per-cohort return is quoted.

  4. STEP 04

    We document the review procedure.

    Quarterly Notion grid, per-cohort threshold indicators and a written reallocation rule: if a cohort LTV drops by X%, here is the move. Handover to your teams or your existing e-commerce agency. You then steer without vendor dependency.

06 · THE FLOW AT A GLANCE

The path each order takes between the first click and the third order.

— not the other way aroundCAC → LTVsequence that drains net marginLTV → CAC LOOPLTVCAC= defendable ratio ≥ 3ACQUISITIONcohorts per channel18 months · entry channelRETENTIONmeasured repurchasewindow per frequencyAVERAGE BASKETcross-sell · up-sellper margin segment++measured at 12 and 24 months per cohortCFO-signed ceiling · per catalog segmentUNIT ECONOMICSdefendable net margindefendable in committeeCOHORT MEASUREMENT STACKGA4 ENHANCED E-COMMERCE · BIGQUERY · LOOKER STUDIO · ERP/SHOPIFYquarterly Notion grid · per-segment signed CAC ceiling

Lifetime value conditions acceptable acquisition cost, never the other way around. Unit economics ship out at the end, defendable at the executive committee.

07 · NOT YET FOR YOU IF

Three cases where laying down an e-commerce strategy is not the top priority.

  • You refuse to share your per-segment margin.

    Without visibility on per-segment gross margin, we cannot calibrate a defendable CAC ceiling. The LTV → CAC mechanic loses its backbone and the plan becomes generic, not actionable.

  • Your e-commerce revenue is below 1 M€.

    Below this threshold, cohort LTV scoping rarely pays back. You will gain more by focusing on two profitable segments than by building a theoretical grid across the full catalog.

  • Your e-commerce tracking layer does not hold.

    Without clean GA4 Enhanced E-commerce and a reliable BigQuery (Google data warehouse) export, LTV modelling rests on estimation. We prefer scoping the measurement layer first before laying down a strategy that will lean on it.

08 · THE QUESTIONS WE ACTUALLY HEAR

Questions whispered after the second meeting. Honest answers.

On the 1-20 M€ e-commerce missions we ran in 2024-2026, the gap between a CAC steered on last-click only and a CAC steered on real customer value (LTV per cohort) sits around 20-35% more net margin on the LTV side. Not net volume, but recovered margin. Prerequisite on the measurement side: a CAC-calibrated media plan that bridges to margin.

Not opposed. Multi-touch attribution (who touched this customer before buying) arbitrates the mix on a short 7-30 day window. LTV cohort tracking validates that a channel truly feeds margin at 12 and 24 months. Without that layer, you over-invest in channels that perform on the platform side but not on the real-value side. Both complement each other. Prerequisite on the collection side: clean server-side tracking that keeps the signal consistent across both windows.

Not to our knowledge. Documented APD (Belgian DPA) sanctions are on tracking without a legal basis, out-of-GDPR email retention and missing Consent Mode (user consent management). Cohort tracking on legally collected and anonymised data is a standard use, comparable to CRM analysis.

Belgian 1-20 M€ e-commerce range we observed: 20-35% more net margin over 18 months (Ecostal Group audit). No guaranteed numbers. The main lever: cut segments above CAC ceiling and reinvest on high-value-over-time cohorts. The marketing automation workflows layer downstream amplifies the return.

The Notion grid costs nothing to maintain. The BigQuery (dedicated data warehouse) + Looker Studio (Google dashboards) layer runs 50-150 € HTVA per month depending on volume. The quarterly review takes 3-5 hours from your team or e-commerce agency. If we leave, you keep the doc, BigQuery queries and dashboards. Pattern documented on the Ecostal Group mission, where the loop has been running without us for 6 months.

Yes, on 30% of e-commerce missions. The rule: who steers the LTV mechanic and who executes acquisition / retention levers. If the existing agency stays operational pilot, we scope the strategy with them. If they execute our frame, we accompany two quarters then step back. We never step on the other side without saying. The broader agency vs freelance vs in-house debate sits here.

Field note

An e-commerce strategy without measured customer value is blind planning. At Ecostal Group, until we extracted the per-acquisition-month data into a dedicated database (BigQuery), we were running on the returns Meta and Google show on their own dashboards. That kind of return is the cherry on top, not the base. At our shop, we start with the export. The rest follows.

Jérôme · Strategy & data
JérômeStrategy & data · HeySquad

Other sub-services STRATEGY

We open your LTV cohorts. We tell you where your CAC ceiling is breached without you seeing it.

Get my e-commerce strategy quotedBack to the Strategy service